Association Of COX-2 Gene Polymorphisms With Susceptibility To Hepatocellular Carcinoma

Part 1 Association of COX-2 gene polymorphisms with susceptibility to Hepatocellular CarcinomaObjectives Cyclooxygenase (COX) is a central enzyme that converts arachidonic acid into prostaglandins. Its isoform COX-2 has a special biological characteristics, which influences carcinogenesis through inflammatory response, apoptosis inhibition and tumor cell invasion, and metastasis. Several potentially functional single nucleotide polymorphism (SNP) sites may affect COX-2 transcriptional acticity and expression levels, which ultimately may affect function of protein. We used hospital-based case-control study to assess the association of COX-2 gene promoter region -1195G>A (rs689466) and 3'UTR 8473T>C (rs5275) SNP with genetic susceptibility to HCC, and the potential interaction of gene-environment on risk of HCC. Methods The study subjects consisted of 780 newly and histopathologically diagnosed cases with HCC and 780 cancer-free controls frequency-matched for age, sex and nationality. A structured questionnaire was administered by interviewers to collect information on demographic data and environmental exposure history. Each subject donated 3mL of peripheral blood, which used for genomic DNA extraction with a Phenol-Chloroform method. Genotypes of COX-2 were detected using Applied Biosystems TaqMan genotyping platform according to the manufacture's recommendations. Hardy-Weinberg genetic equilibrium (HWE) for all two SNPs was tested in controls and Haplotype analysis using SHEsis software (http://analysis.bio-x.cn/SHEsis Main.htm). Differences in the distributions of demographic characteristics and environmental exposure between the cases and controls were evaluated using the Student's t test (for continuous variables) and Chi-square test (for categorical variables). Odds ratio (OR) and 95% confidence interval (CI) were obtained from unconditional univariate and multivariable logistic regression model to evaluate associations between all two SNPs and HCC risk. All the statistical analyses were performed with SSPS software (version 13.0), all tests were two-sided withαis 0.05.Results⑴Demographics and risk factors for study subjects There were no statistically differences in the distributions of age, sex, and nationality between cases and controls (P>0.05), respectively. However, there were more ever smokers, ever alcohol users and ever HBV infection individuals among the cases than among the controls, and these differences were statistically significant (P<0.01).⑵Association of COX-2 SNP with the risk of HCC The genotype frequency distributions of COX-2 -1195G>A (rs689466) and 8473T>C (rs5275) were in agreement with the HWE among the controls (P>0.05). The genotype and allele frequencies of all two polymorphisms were not statistically different between cases and controls (P>0.05), respectively. No significant association was observed between the GA genotype and the AA genotype of COX-2 -1195G>A compared with the GG genotype (adjusted OR = 1.30, 95%CI = 0.91-1.86, P = 0.147; adjusted OR = 1.35, 95%CI = 0.89-2.05, P = 0.152, respectively) or the TC genotype and the CC genotype of COX-2 8473T>C compared with the TT genotype (adjusted OR = 0.85, 95%CI = 0.62-1.16, P = 0.308; adjusted OR = 1.10, 95%CI = 0.51-2.38, P = 0.803, respectively) and the risk of HCC by genotype alone,adjusted with the factors of age, sex, HBV infection, drinking, smoking, and family history of cancer.We further performed stratified analysis by the factors of age, sex, HBV infection, drinking and smoking status. We found that under a dominant model the -1195G>A GA+AA genotypes increased the risk of HCC among aged < 55 years subjects (adjusted OR = 1.56, 95%CI = 1.03-2.37, P = 0.037), 8473T>C TC+CC genotypes decreased the risk of HCC among female subjects (adjusted OR = 0.50, 95%CI = 0.25-0.99, P = 0.047). Interaction between 8473T>C variant and sex was found to be of borderline significance (adjusted OR = 0.48, 95%CI = 0.23-1.03, P = 0.060), we found no statistical evidence for interactions between -1195G>A variant and age in the multivariate logistic regression model (adjusted OR = 1.58, 95%CI = 0.78-3.23, P = 0.207)There was not significant linkage disequilibrium (LD) between COX-2 -1195G>A and 8473T>C (r~2 = 0.192). Haplotype analysis revealed that three common haplotypes (>5%) A-1195T8473, G-1195T8473, G-1195C8473 were not associated with the risk of HCC (OR = 0.97, 95%CI = 0.85-1.12, P = 0.716; OR = 1.05, 95%CI = 0.90-1.22, P = 0.537; OR = 0.98, 95%CI = 0.82-1.17, P = 0.785).Conclusions COX-2 -1195G>A GA+AA genotypes may increase the risk of HCC among aged < 55 years subjects, 8473T>C TC+CC genotypes may decrease the risk of HCC among female subjects. Interaction between 8473T>C variant and sex was found to be of borderline significance, we found no statistical evidence for interaction between -1195G>A variant and age. Larger sample size studies are warranted to further confirm our findings. Objectives Cyclooxygenase (COX) is a central enzyme that converts arachidonic acid into prostaglandins. Its isoform COX-2 has a special biological characteristics, which plays an important role in the study of cancer etiology. A number of studies have been conducted to examine the association between COX-2 -1195G>A and 8473T>C polymorphisms with the susceptibility to cancer, but the results remain controversial. We conducted a Meta-analysis to evaluate the association between these two polymorphisms and cancer susceptibility.Methods Pubmed, EMBASE, Springerlink, PML, CNKI, VIP and Wanfang Data were searched using the search terms:"COX-2","Cyclooxygenase2","PTGS2","polymorphism","variants","Neoplasms","cancer","Huan yang hua mei2(in Chinese)","Duo tai xing(in Chinese)"and"Zhong liu(in Chinese)". All relevant publications were reviewed, including heterogeneity test, subgroup analysis, sensitivity analysis and publication bias investigation. All selected studies were no related case-control study, for which the genotype frequency distributions were in agreement with the HWE among the controls. All the statistical test for this Meta-analysis were performed with software Review Manager version 5.0 and Stata version 10.0.Results⑴Study characteristics There are 30 case-control studies with 9973 cancer cases and 12917 controls for cancer susceptibility related to the -1195G>A polymorphism including 6 Chinese and 24 English studies and 34 case-control studies with 22859 cancer cases and 28136 controls for cancer susceptibility related to the 8473T>C polymorphism including 3 Chinese and 31 English studies.⑵Association of COX-2 SNP with the risk of Cancer COX-2 -1195G>A polymorphism can increase the risk of cancer (dominant model OR = 1.13, 95%CI = 1.04-1.23; recessive model OR = 1.12, 95%CI = 1.03-1.23); 8473T>C polymorphism seemed to be not associated with the risk of cancer in any model (dominant model OR = 0.99, 95%CI = 0.93-1.05; recessive model OR = 1.05, 95%CI = 0.96-1.15).⑶Subgroup analysis①In the subgroup analysis stratified by ethnicity, COX-2 -1195G>A GA or AA genotype increased the risk of Asians cancer, compared with the GG genotype (OR = 1.12, 95%CI = 1.01-1.23; OR = 1.35, 95%CI = 1.13-1.62). 8473T>C polymorphism seemed to be not associated with the risk of Asians cancer (dominant model OR = 0.96, 95%CI = 0.88-1.04).②In the subgroup analysis stratified by tumor type, COX-2 -1195G>A polymorphism may be not associated with the risk of HCC (dominant model OR = 1.14, 95%CI = 0.95-1.36; recessive model OR = 1.20, 95%CI = 0.87-1.66). 8473T>C TC+CC genotype decreased the risk of bladder cancer, compared with the TT genotype (OR = 0.75, 95%CI = 0.63-0.90). COX-2 SNPs may be not associated with the risk of Colorectal cancer, Esophageal cancer, Lung cancer, Prostate cancer, or Breast cancer in subgroups of tumor type.③In the subgroup analysis stratified by control source and sample size, AA genotype of -1195G>A can increase the risk of cancer in subgroups of population-based studies or a larger sample size (controls based on population, OR = 1.26, 95%CI = 1.03-1.55; sample size 500-1000, OR = 1.31, 95%CI = 1.11-1.56; sample size >1000, OR = 1.17, 95%CI = 1.02-1.34), 8473T>C polymorphism seemed to be not associated with the risk of cancer in subgroups of population-based studies or a larger sample size.④In the subgroup analysis stratified by matching status, COX-2 -1195G>A polymorphism can increase the risk of cancer in subgroups of matching design (dominant model OR = 1.11, 95%CI = 1.01-1.21; recessive model OR = 1.11, 95%CI = 1.01-1.21), In subgroups of no-matching design, we only found -1195G>A polymorphism can increase the risk of cancer in dominant model (OR = 1.37, 95%CI = 1.06-1.76).⑤In the subgroup analysis stratified by genotyping, COX-2 -1195G>A or 8473T>C polymorphism may be not associated with the risk of cancer in subgroups of PCR-RFLP or TaqMan.However, lacking of the original data of the reviewed studies limited our further evaluation, We failed to perform subgroup analysis by the factors of age, sex, drinking and smoking status on COX-2 polymorphisms and cancer susceptibility.Conclusions COX-2 -1195G>A polymorphism may be associated with the risk of cancer, ethnicity may be the source of heterogeneity. 8473T>C polymorphism may be not associated with the risk of cancer, but subgroup analysis revealed 8473T>C polymorphism was associated with the risk of bladder cancer, tumor type and study design may be the source of heterogeneity.